(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Performance assessment of the tension infiltrometer."

Centre for Land 
and Biological Resources Research 



vy^ 



4v 



Centre de recherches sur les 
terres et les ressources biologiques 




<K 



PERFORMANCE ASSESSMENT OF 
THE TENSION INFILTROMETER 



M-M, U Agriculture 



Canada 






} 



Library / Bibliotheque, Ottawa K1A 0C5 




Marketed by 

Soilmoisture Equipment Corp., 

Santa Barbara, California 

Research Branch 
Technical Bulletin 1 993-1 2E 



1*1 



6 Jo 7J? 

c 



Agriculture 
Canada 

Research Direction generale 
Branch de la recherche 



Canada 



I 

I 
I 
I 
[ 
r 

i 

i 
i 
i 

i 

i 

i 
i 
i 
i 

i 



Performance Assessment of the 
Tension Infiltrometer 



Marketed by Soilmoisture Equipment Corp., 
Santa Barbara, California 



by 

W.D. Reynolds and W.D. Zebchuk 

Centre for Land and Biological Resources Research 

Research Branch, Agriculture Canada, 

Ottawa, Ontario 



Technical Bulletin 1993-12E 



Centre for Land and Biological Resources Research 
Research Branch, Agriculture Canada 

May 1993 



Copies of this publication can be obtained from: 

W.D. Reynolds 

Centre for Land and Biological Resources Research 

Research Branch, Agriculture Canada 

Central Experimental Farm 

Ottawa, Ontario 

K1A 0C6 

Telephone: (613) 995-5011 

Fax: (613) 995-1823 

E-Mail: AC230SOIL@NCCCOT2.ACR.CA 

Published by Cartographic Design and Reproduction Unit 
Centre for Land and Biological Resources Research 
CLBRR Contribution 93-50 

© Minister of Supply and Services Canada 1993 
Cat. No. A54-8/1993-12E 
ISBN 0-662-21006-9 



Contents 

1. Introduction 1 

2. Wettability and Bubbling Pressure of the Porous Disk . . 3 

3. Hydraulic Conductivity of the Porous Disk 3 

4. Porous Disk Leakage Problems 5 

5. Bubble Tower Calibration 5 

6. Maintenance of the Porous Disk 10 

7. Comments Concerning Field Operation 10 

8. Recommendations 12 

References Cited 13 

Tables 

Table 1 4 

Table 2 8 

Figures 

Figure 1 2 

Figure 2 6 

Figure 3 7 

Figure 4 9 



in 



Digitized by the Internet Archive 

in 2013 



http://archive.org/details/performanceasse199312cent 



1. Introduction 



The tension infiltrometer (TI) is a field apparatus for in-situ 
determination of near-saturated hydraulic conductivity, K(h), 
where "near-saturated" refers to measurements made over the 
tension or negative pressure head (h) range, j< h j< 20 cm of 
water. The apparatus operates by measuring the infiltration 
rate (recharge) into the soil through a porous disk or 
membrane on which a constant water tension is applied by a 
"bubble tower" (Fig. 1). The K(h) values are obtained by 
measuring a sequence of infiltration rates corresponding to a 
sequence of tensions set on the bubble tower, and then 
applying theoretical relationships given in Reynolds (1993), 
Ankeny (1992), Elrick and Reynolds (1992) and White et al. 
(1992). 

Tension infiltrometers have just recently moved from the 
realm of "research prototype" to "commercial product", and 
are currently marketed by A.L. Franklin Pty. Ltd., Sydney, 
Australia; Soil Measurement Systems, Tucson, Arizona; and 
Soilmoisture Equipment Corporation, Santa Barbara, 
California. The information available on performance 
attributes and maintenance is somewhat limited for all of these 
"commercial" TI's, but particularly so for the unit marketed by 
Soilmoisture Equipment Corp. Specifically, Soilmoisture 
provides no information on the hydraulic properties and 
maintenance of the porous disk, or on the calibration of the 
bubble tower. As detailed knowledge of these attributes is 
critical to the successful operation of tension infiltrometers, 
this performance assessment of the Soilmoisture TI was 
undertaken to establish the wettability, bubbling pressure and 
hydraulic conductivity of the porous disk (disk hydraulic 
properties), and the accuracy, linearity and resolution of the 
bubble tower (bubble tower calibration). Brief comments and 
recommendations are also given concerning maintenance of 
the porous disk, and operation of the TI in the field. A total 
of 10 Soilmoisture TI's were assessed. Note that the 
Soilmoisture TI is occasionally referred to as the "Guelph 
Tension Infiltrometer" (GTI) because it is designed as an 
attachment to their "Guelph Permeameter", which they have 
marketed for several years. 



Reservoir Air 
Tube (Blocked) 



Reservoir 



Water 

Supply 

Tube 




Transparent Polycarbonate 
Infiltrometer Plate 
(10-20 cm diameter) 



_ Retaining 
Ring 



Soil 
surface 



Porous Disk/ 
Membrane 



.270 Mesh Nylonl 
Bolting Cloth 



Figure 1 . Schematic of a mariotte-based tension infiltrometer 
(after Reynolds, 1993). 



2. Wettability and Bubbling Pressure of the 
Porous Disk 

The porous disk or membrane in any TI (Fig. 1) must be 
hydrophillic (water wettable) and have a distinct bubbling 
pressure (air entry value) that is greater than the maximum 
tension to be applied by the bubble tower. The porous disks 
in the Soilmoisture TTs did not wet spontaneously when first 
placed in de-aired, temperature-equilibrated tap water, even 
when left standing in the water for several days. The disks 
did wet, however, when vacuum was applied. The bubbling 
pressure of the disks was found (using the procedures in 
Reynolds, 1993) to be approximately 30-35 cm of water, which 
exceeds (as is required) the maximum tension that can be 
applied by the bubble tower (approximately 25 cm of water). 

It is desirable for the porous disks to wet spontaneously (i.e. 
without applying a vacuum) when placed in water, as this 
produces a greater degree of saturation of the disk and 
consequently a greater disk conductivity (discussed further in 
Section 3). The disks in the Soilmoisture TTs consist of porous 
polyethylene plastic that has been treated with a chemical 
surfactant to induce wettability (polyethylene is naturally 
hydrophobic, i.e. non-water wetting). We found that their 
wettability can be improved to the point of spontaneous 
wetting by submerging the disks in a surfactant solution (49% 
by vol. isopropyl alcohol, 49% by vol. distilled water, 2% by 
vol. Triton X 100 surfactant) for 24-36 hours. 



3. Hydraulic Conductivity of the Porous Disk 

The saturated hydraulic conductivity (Ksat) of the porous disk 
should be greater than the field-saturated and near-saturated 
hydraulic conductivity of the porous material being tested. 
Otherwise, the disk may restrict water infiltration, which may 
in turn result in underestimates of the conductivity of the 
porous material. The Ksat values of the Soilmoisture disks 
after initial saturation are given in Table 1. The mean Ksat of 
1.30 x 10~ 3 {+_ 0.248 x 10~ 3 ) cm/s is about an order of magnitude 
lower than what one might consider as ideal, since the field- 



Table 1. Saturated hydraulic conductivity (Ksat) of the Soilmoisture Tl 
porous disks after initial saturation from an air-dry state. 



Tl Ident. No. 


10 3 Ksat + (cm/s) 


1 


0.896 


2 


1.70 


3 


1.40 


4 


1.33 


5 


1.51 


6 


1.62 


7 


1.18 


8 


1.19 


9 


1.10 


10 


1.04 


Arithmetic Mean 


1.30 


Standard Deviation 


0.248 



+ measured using the falling head method 



saturated hydraulic conductivity of most agricultural soils is 
_< 10~ 2 cm/s. When the disks were treated with the surfactant 
solution, the mean Ksat increased by about a factor of 2.6 to 
3.32 x 10~ 3 (+ 0.292 x 10 -3 ) cm/s. This increase apparently 
reflects a greater degree of disk saturation resulting from the 
increased wettability. When some of the disks were re-wetted 
after about 3 months storage in an air-dry state, the mean Ksat 
had dropped to 1.65 x 10 -3 (+ 0.169 x 10~ 3 ) cm/s, which may 
suggest that the disks require periodic re-treatment with 
surfactant (say, every 2-4 months) in order to maintain 
maximum Ksat. 

Although the Ksat of Soilmoisture's porous disk is lower than 
ideal, it should still be quite adequate for most agricultural 
soils, particularly for measuring the near-saturated 
conductivities which are often more than an order of 
magnitude less than the field-saturated value. As discussed 
above, K(h) results that are greater than or equal to the Ksat of 
the porous disk (i.e. greater than about 3 x 10 -3 cm/s) should 
be treated with caution. The minimum hydraulic conductivity 



that can be measured conveniently by Soilinoisture's TI system 
appears to be about lO^ 6 cm/s, which gives the TI a K(h) range 
of about 3 x 10~ 3 cm/s to 1 x 10 -6 cm/s. The minimum K(h) is 
determined primarily by the bubbling pressure of the porous 
disk/contact sand, and the minimum infiltration rate that can 
be measured accurately within a practical period of time. 



4. Porous Disk Leakage Problems 

The seal between the porous disk and the aluminum support 
ring (Fig. 2) must be air-tight to prevent air leaks when 
tension is applied to the disk. This seal, which is made with 
glue in the Soilmoisture design, failed in several of the units 
when measuring the Ksat of the porous disk. The problem 
appeared to be the inability of the glue to grip the anodized 
aluminum support ring. All alternative glues tried (including 
epoxies such as Conap) also failed to grip the aluminum 
adequately. This situation was resolved by machining a step 
into the porous disk and attaching an aluminum "retaining 
ring" (Fig. 3). A bead of silicone (G.E. silicone "gasket maker") 
was placed between the porous disk and support ring, and 
between the porous disk and retaining ring (Fig. 3). The 
silicone forms an air-tight seal, but still allows the disk to be 
easily removed if required. The retaining ring supplies the 
mechanical force required to hold the porous disk in place and 
to maintain the integrity of the seal. The porous disks were 
re-treated with wetting agent after the installation of the 
retaining rings in order to make the surfaces of the silicone 
seal water wettable. 



5. Bubble Tower Calibration 

The accuracy, linearity and resolution of the TI bubble tower 
for setting the tension on the porous disk was determined 
using a tension table-hanging water column arrangement, and 
procedures similar to those given in Reynolds (1993). In 
essence, the TI's were placed on the tension table, a range of 
tensions were set on the bubble towers, and the resulting 
steady state tensions on the porous disks were read via the 
hanging water column. The accuracy and linearity of the 
bubble towers were assessed by determining the least squares 



Vacuum ^2\ 

Port L_ ZJ 



Bubbler 
Top Cap 

Bubbler 
Rubber 
Seal 



Porous Disk 



Air Tube & 
Rubber Seal 



Guelph 

Reservoir 

Assembly 




Port Plug for Pressure 
Transducer Port 



Air Inlet Tip 



Support Ring 



Figure 2. Schematic of the tension infiltrometer marketed by Soilmoisture 
Equipment Corp. (after Soilmoisture Equipment Corp., Santa 
Barbara, California). 



Marriot 

Bubbler 

Assembly 




Figure 3. Schematic of the aluminum retaining ring attachment for the 
Soilmoisture Equipment tension infiltrometer. 



regression relationship between the tension set on the bubble 
tower (Y-axis) and the tension measured by the hanging water 
column (X-axis). The tensions used were approximately, 24, 
15, 7 and 1 cm of water, which effectively covers the full range 
of tensions available on the bubble tower. The resolution of 
the bubble towers was estimated by observing the oscillation 
in measured tension over time (as determined by hanging 
water column) for particular bubble tower tensions. 

The least squares regression coefficients and R-squared values 
for each TI are given in Table 2. A summary plot of bubble 
tower tension vs. hanging water column tension for all 10 TI's 
is given in Fig. 4, along with the corresponding regression line 
and the 1:1 line. It is seen in Table 2 that the regression 
relationships are very similar and highly linear, with a mean 
slope of 1.0001 (+ 0.0053), a mean intercept of -1.6548 
(+ 0.1818) cm, and a mean R-squared of 0.99997 (+ 0.0000293). 
This is confirmed in Fig. 4, where the regression slope, 
intercept and R-squared values are 0.9998, -1.6514 and 0.9997, 
respectively. The bubble towers are consequently very similar 



Table 2. Least squares regression relationships between set bubble 

tower tension (Y) and measured hanging water column tension 
(X), Y = aX + b, and the corresponding coefficients of 
determination, R-squared. 



TI Ident. No. 


a 


b (cm) 


R-Squared 


1 


1.0102 


-1.8360 


0.99997 


2 


1 .0001 


-1 .5394 


0.999898 


3 


0.9878 


-1 .2620 


0.99996 


4 


1 .0023 


-1 .4928 


0.99998 


5 


0.9983 


-1 .5640 


0.999997 


6 


0.9980 


-1 .7349 


0.999998 


7 


1 .0035 


-1 .7846 


0.99997 


8 


1 .0000 


-1 .6500 


1 .0000 


9 


0.9989 


-1.8107 


0.99998 


10 


1.0017 


-1 .8735 


0.99995 


Arithmetic Mean 


1.0001 


-1 .6548 


0.99997 


Standard Deviation 


0.0053 


0.1818 


0.0000293 



8 



E 

c 
g 

c 

CD 

|2 

-D 
.Q 

13 
CO 



30 - 














X 


Measured values 


(all 10 units) 








■ • 


■ Overall Regression Line 






25- 












20 - 










^ • 

^ ♦ 

^ ♦ 
* 

* 
• 


15- 












10 - 








* 


Regression Equation 


- 










Y = .999821X- 1.65135 
R-Square = 0.999689 


5 - 
n 




^ • 
^ ♦ 








U 


i 


i i i i r 


1 I | I II 


1 III 


i I i i i i i i i i 



5 10 15 20 25 

Hanging Water Column Tension (cm) 



30 



Figure 4. Summary plot of bubble tower tension versus hanging water 

column tension for all 10 Soilmoisture Tl units. The crosses are the 
individual measurements and the broken line is the regression 
through the data. The solid line is the 1:1 line. 



in their response, and this response is highly linear over the 
full range of tensions. The non-zero regression intercepts 
indicate, however, that the bubble towers underestimate the 
actual tension on the porous disk, by an average of 1.65 cm. 
That is, the actual tension on the base of the porous disk is, on 
average, 1.6548 (+ 0.1818) cm greater than indicated by the 
bubble tower (Table 2). Probably the best way to account for 
this "offset" is to place a 1.65 cm layer of contact sand between 
the porous disk and the soil (Fig. 1). This makes the tension 
at the soil surface equal to the tension indicated by the bubble 
tower, and simultaneously ensures good hydraulic contact 
between the porous disk and the soil. Recommendations 
regarding the type of material to use as a contact sand and 
how it might be placed are given in Reynolds (1993). 



The hanging water column tensions tended to oscillate 
approximately + 0.1 cm around the steady values. This 
oscillation was in phase with the bubbling of the bubble tower, 
and was therefore probably the result of pressure pulses 
caused by the breaking bubbles. The laboratory resolution of 
the bubble towers therefore appears to be about +_ 0.1 cm. 
This resolution probably drops to around + 0.5 cm in the 
field, due to additional disturbances such as wind, solar 
heating, irregularities in the soil surface, and the accuracy with 
which the contact sand can be placed. 



6. Maintenance of the Porous Disk 

Over a field season of continuous use (say, 4 months), the 
Ksat of the porous disk may gradually decline due to the 
progressive accumulation of algal/mould growths (these 
appear as green and grey patches on the disk) and 
iron/aluminum precipitates (red and brown patches). These 
accumulations can be removed by soaking the disk for about 
24 hours in concentrated bleach solution (30% by vol. 
commercial bleach, 70% by vol. distilled water) to remove the 
algae and moulds; and then soaking again (after flushing out 
the bleach) for about 48 hours in dithionite-citrate solution 
(0.4 g dithionite per 25 ml of 0.68 M sodium citrate solution) to 
remove the iron/aluminum precipitates. This treatment also 
removes the chemical wetting agent, however, and the disk 
must consequently be re-treated with surfactant to 
re-establish its wettability. The disk may have to be removed 
from its aluminum support and retaining rings (Fig. 2, 3) 
before cleaning, as bleach tends to react with unanodized 
aluminum. 



7. Comments Concerning Field Operation 

On some of the TI units, the support tube tended to pull out 
of the foot cover when the TI was full of water due to 
slippage of the O-ring connection (Fig. 2). It would 
consequently be advisable to develop a means for locking this 
connection when using the TI in the field. It is also felt that 



10 



the bubble tower (i.e. marriot bubbler assembly, Fig. 2) should 
be protected with some form of brace or shield during field 
operation, as its connection to the foot cover appears 
somewhat fragile. 

If the TI is not level, air bubbles from the marriot bubbler 
assembly (bubble tower) may accumulate under the foot cover, 
rather than entering the support tube and rising up into the 
reservoir (Fig. 2). This situation should be avoided, as it 
introduces error into the infiltration rate measurements. 



8. Recommendations 

Although only laboratory testing of the Soilmoisture TI's has 
been performed at this writing, it nevertheless appears that 
they will perform well in the field providing that certain 
precautions are taken and procedures followed. These 
include: 

i) periodically treating the porous disks (perhaps every 2-4 
months) with a chemical surfactant to establish and 
maintain maximum wettability and Ksat. 

ii) periodically cleaning the porous disks (perhaps every 

4 months of continuous use) to remove growths and 
precipitates and thereby re-establish maximum Ksat. 

iii) not attempting to use the TI outside of its K(h) range 
(approx. lOr 6 cm/s < K(h) < 3 x 10~ 3 cm/s). 

iv) periodically testing for air leaks in the seal between the 
porous disk and the support ring, and installing a 
retaining ring (Fig. 3) if necessary. 

v) compensating for the 1.65 cm tension offset 

(underestimate) of the bubble tower, preferably by 
placing a 1.65 cm thick layer of contact sand between 
the porous disk and the soil surface. 

vi) following the general procedures and guidelines for 
field use of TTs recommended by Reynolds (1993), 
Ankeny (1992), Elrick and Reynolds (1992) and White 
et al. (1992). 



11 



References Cited 

Ankeny, M.D. 1992. Methods and theory for unconfined infiltration 

measurements, pp. 123-141. In G.C. Topp et al. (ed.) Advances 
in measurement of soil physical properties: Bringing theory 
into practice. SSSA Spec. Publ. 30. Soil Science Society of 
America, Madison, Wis. 

Elrick, D.E. and W.D. Reynolds. 1992. Infiltration from constant-head 
well permeameters and infiltrometers. pp. 1-24. In G.C. Topp 
et al. (ed.) Advances in measurement of soil physical 
properties: Bringing theory into practice. SSSA, Spec. Publ. 30. 
Soil Science Society of America, Madison, Wis. 

Reynolds, W.D. 1993. Unsaturated hydraulic conductivity: Field 

measurement, pp. 633-644. In M.R. Carter (ed.) Soil sampling 
and methods of analysis. Canadian Society of Soil Science, 
Ottawa, Ont. 

White, I., M.J. Sully, and K.M. Perroux. 1992. Measurement of surface- 
soil hydraulic properties: Disc permeameters, tension 
infiltrometers and other techniques, pp. 69-104. In G.C. Topp 
et al. (ed.) Advances in measurement of soil physical 
properties: Bringing theory into practice. SSSA Spec. Publ. 30. 
Soil Science Society of America, Madison, Wis. 



13 



,/ANAl»l 



.-Jilll,. 

3 ^073 00l017b0 T